Cooperative Estimation of Maps of Physical and Virtual Radio Transmitters

Abstract

In multipath assisted positioning, the spatial information contained in multipath components (MPCs) is exploited, as MPCs are regarded as line-of-sight signals from virtual transmitters. The positions of physical and virtual transmitters can be estimated jointly with the receiver position with simultaneous localization and mapping (SLAM). In our multipath assisted positioning approach called Channel-SLAM, the estimates from a channel estimator are used in a Rao-Blackwellized particle filter which implements SLAM. While the original Channel-SLAM algorithm is a single-user positioning system, we present a comprehensive framework for cooperative Channel-SLAM within this paper. Users cooperate by exchanging maps of estimated transmitter locations. With prior information about the locations of physical and virtual transmitters, the positioning performance of the users increases significantly. The more users contribute to such a transmitter map, the more increases the positioning performance. With simulations in an indoor scenario, we show that the positioning performance is bounded for cooperative Channel-SLAM in the long run.